In the early 1970's, a great deal of research was done on synthesized speech. The need for this was in the area of computerized systems that could actually provide audible information in the form of commands, directions, etc. to a listener. This technology found a use in the game market and such companies as Texas Instruments developed algorithms for generating speech, such as the linear predictive coding (LPC) algorithm, a technique for synthesizing audible speech patterns. At this time, memory was quite expensive and the density thereof was inadequate to provide for storage of prerecorded information that was digitized, so a hardware based algorithm was more practical. Some of the early integrated circuits that provided for the output of audible sounds through LPC based algorithms involved such things as “talking greeting cards” wherein a chip and associated battery with an integrated speaker were disposed within a greeting card such that, when the greeting card was opened, a greeting was provided. Some of the original greeting cards had “canned” greetings. With the advent of technology, audible files have been compressed in what is termed as a “WAV” file such that music and voice can be transferred over computer networks such as packet-switched networks. However, one of the limiting factors to incorporating these WAV files into small integrated circuits or hybrid circuits is the requirement for the memory to store the information that is to be played back and the ability to adaptively record such information.